New model allows doctors to predict lung cancer in smokers with 85% accuracy

Scientists have been able to accurately predict whether a person is likely to develop lung cancer using only three data points using a revolutionary computer model according to a new study reported by EurekAlert.

The study was conducted by Thomas Callender from University College London alongside a team of researchers and published in the peer-reviewed journal PLOS Medicine.

The computer model uses the number of packets smoked per day, the number of years the individual smoked for, and their age. 

The scientists hope that the new system will allow doctors to better advise patients who may be at higher risk to undergo lung cancer screenings. 

Lung cancer remains the most common cause of death worldwide and has a low survival rating for those who fail to catch the problem early on.

Unfortunately, it has been difficult for doctors to know who to prioritize for such screening, with previous standards requiring up to 17 different bits of patient information, not all of which is available at a routine checkup.

Leagues ahead

The new computer model designed by the researchers was able to predict both the possibility of contracting lung cancer and the likelihood of death with almost 85% sensitivity.

Callender said of the study, “We know that screening for those who have a high chance of developing lung cancer can save lives. With machine learning, we’ve been able to substantially simplify how we work out who is at high risk, presenting an approach that could be an exciting step in the direction of widespread implementation of personalized screening to detect many diseases early.”